LIBRARY

Michigan State
University

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

MTE DUE

MTE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1m mimmd4

EFFECT OF HEAT TREATMENT AND SURFACE CHARGE
HETEROGENEITIES ON THE INITIAL RATE OF
FLOCCULATION OF POLYSTYRENE PARTICLES

by

Comic Jo Leuderalbert

A THESIS

Submitted to
Michigan State University
in partial fiJlfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Chemical Engineering

1 997

ABSTRACT

EFFECT OF HEAT TREATMENT AND SURFACE CHARGE
HETEROGENEITIES ON THE INITIAL RATE OF
F LOCCULATION OF POLYSTYRENE PARTICLES

By
Connie Jo Leuderalbert
DLVO theory, which governs particle-particle interactions, has been shown to predict a
steeper stability ratio curve than is obtained experimentally during slow flocculation of
colloidal particles. The objective of this work is to examine the effect of two of the more
likely reasons for this classical discrepancy: surface roughness and surface charge

heterogeneities.

Surfactant fi'ee polystyrene beads of D = 0.30 pm were obtained from lnterfacial Dynamics
corporation for the study. To reduce surface rouglmess, dispersions containing 0.05% solids
were heated in a Parr bomb for 6, l2 or 24 hours at 120°C followed by slow cooling to
examine the effect of length of heat treatment. Surface charge heterogeneities were
mitigated by adding sodium dodecyl sulfate to obtain a final dispersion concentration of

mo" M.

Results show that the effects of heat treatment are fully realized by 6 hours of heating at
120°C. It was also found that the addition of a surfactant lessens the disagreement between

theory and experiment.

DEDICATION

TO

The loving memory of my father
who taught me that anything is possible

Mom
whose constant support encourages me to excel

Gary, Sandy, Jeff, Stephanie, Sharon, Jim, James, Scott, Ken, Jody, Rachel, Megan,
Sean and Jessica
for your love and support, and for providing me with so much joy

Jody
for always being there for me, even in the bad times

Steve
for letting me lean on him when it all seemed hOpeless
and for teaching me to believe in the future
and

Abhijit
for reminding me just how lucky I am; I wish life had given you more time

iii

ACKNOWLEDGMENTS

Thanks to my professor and friend Dr. Bob Ofoli for his help and guidance throughout my
graduate career. Not only was he there for any technical questions I had, he was there
whenever I needed to talk.

Special thanks also goes to my thesis committee, Dr. Bob Ofoli, Dr. Alec Scranton and Dr.
Dennis Miller, for taking the time out of their busy schedule.

Thanks to my groupmates Ari, Pari, Sumei, Abhijit, Jamaica, Brian, Bill, Avani and Jeff for
making work fim and for helping me to have fun in my spare time.

Thanks to the wonderful staff in the Chemical Engineering Department (Julie, Faith and
Candy) for always being willing to help... .even when I was just too lazy to do things myself.

Thanks to the Food Science staff, especially Sharon, for their help and guidance.

Thanks to all the other graduate students and friends that I made during my MSU career,
especially Cassandra, Shawn, Caroline, Julie, Tim, Rob and Scott. They listened to all my
complaints and STILL liked to hang out with me.

Thanks to my roommates Caroline and Tyler for putting up with me and for making me
laugh so often.

Thanks to Bill Schmitt for setting up my light scattering equipment.

Special thanks to the companies and people who so selflessly let me use their equipment and
time: Anna Morfesis at PPG, and John Klier, Yohannes Chonde, and Jason Vallence at Dow.

iv

TABLE OF CONTENTS

 

 

 

 

 

 

 

LIST OF TALLES vii
LIST OF FIGURES viii
NOMENCLATURE ix
1. INTRODUCTION 1
1.1 STABILITY RATIO ................................................................................................................ 2
1.2 PROBLEM DEFINITION ......................................................................................................... 2
1.2.1 Modification Of Surface Morphology By Heat Treatment ....................................... 4

1.2.2 Reduction Of Surface Charge Heterogeneity By Addition Of A Surfactant ............ S

2. GENERAL BACKGROUND 7
2.1 DLVO THEORY ................................................................................................................... 7
2.2 THE EXPERIMENTAL STABILITY RATIO .............................................................................. 8
2.3 THE THEORETICAL STABILITY RATIO ................................................................................ 9
2.3. 1 . 1 Homo-flocculation ............................................................................................... 9

2.3.1.2 Heteroflocculation ............................................................................................... 9

2.4 SMALL ANGLE LIGHT SCATTERING .................................................................................. 10
2.5 ELECI'ROPHOREI‘IC MOBILITY .......................................................................................... 12

3. OBJECTIVES l4
4. MATERIALS AND METHODS 16
4.1 EXPERIMENTAL DESIGN .................................................................................................... 16
4.2 SAMPLE PREPARATION ..................................................................................................... 16
4.2.1 Heat Treatment of Colloidal Dispersions ................................................................ 16
4.2.2 Stabilimtion of Dispersions with Sodium Dodecyl Sulfate (SDS) ........................ 18

4.3 SMALL ANGLE LIGHT SCATTERING ................................................................................... 18
4.3.1 Light Scattering Setup ............................................................................................... 18
4.3.2 Necessity of Clean Experiments ............................................................................... 21
4.3.3 Determination of Single Scattering Limit ................................................................ 23
4.3.4 Initial Flocculation Rate Experiments ...................................................................... 23

V

5. RESULTS AND DISCUSSION 26

 

 

 

 

 

 

5.1 SINGLE SCATTERING LIMIT ............................................................................................... 26
5.2 ELECTROPIIORESIS ............................................................................................................ 28
5.3 ELECTROPHOREI‘IC MOBILITY .......................................................................................... 28
5.4 CRrrICAL COAGULATION CONCENTRATION .................................................................... 33
5.5 THE STABILITY RATIO ....................................................................................................... 33
5.5.1 Effect Of The Extent Of Heat Treatment. ................................................................ 36
5.5.2 Effect of Stabilization with Sodium Dodecyl Sulfate ............................................. 40

5.6 DIFFICULTY INCHARACTERIZING SLOPES OF SCATTERING INTENSITY PROFILES ......... 42
5.7 COMPARISON OF T‘IIEOREIICALANDEXPERIMENTAL STABILITY RATIOS ...................... 44
6. SUMMARY AND CONCLUSIONS 49
6.1 SUMMARY ......................................................................................................................... 49
6.2 CONCLUSIONS ................................................................................................................... 51

7. SUGGESTIONS FOR FUTURE RESEARCH 52
8. APPENDIX A: STOCK SOLUTION CALCULATIONS 53
9. APPENDIX B: LABVIEW 55
10. APPENDIX C: RESULTS .56
BIBLIOGRAPHY 62

 

Table 4-1:

Table 5-1:

Table 5-2:

Table 5-3.

Table 5-4:

Table 8-1:

T able 8-2:

Table 8-3:

Table 8—4:

Table 8-5:

LIST OF TABLES

 

 

 

 

 

 

 

 

 

 

 

Experimental Design I 7
Relative reduction in stability ratio upon heating, % 36
T wical deviation in experimental stability ratio ‘43
Comparison between theoretical and experimental stability ratios 44
Unstabilized particles 47
Concentration calculator for stock colloid 53
Calculator for stock SDS solution 53
Calculator for high molarity K Cl stock solution 53
Calculator for low molarity K Cl stock solution 54
Worksheet used for typical colloidal dispersions 54
Table 10-1: Stability ratio data sheet for native colloidal dispersion 56
Table 10-2: Stability Ratio data sheet for stabilized colloidal dispersion 5 7
Table 10-3: Stability ratio data for 24 hour heat treated native dispersion 58

 

Table 10-4: Stability ratio data for 24 how heat treated stabilized colloidal dispersion___59

Table 10-5: Stability ratio data for 6 and 12 hour heat treated native and stabilized colloidal

dispersions

 

64

 

Table 10-6: Zeta Potentials

vii

_6l

Figure 1-1:
Figure 2-1:
Figure 4-1:
Figure 4-2:
Figure 5-1 :
Figure 5-2:
Figure 5—3:
Figure 5-4:
Figure 5—5:
Figure 5-6:
Figure 5 - 7: Effect of Heat Treatment on Polystyrene particles
Figure 5-8:
Figure 5-9:
Figure 5-10: Normalized stability ratio

Figure 9-1:

LIST OF FIGURES

Typical stability ratio curve

 

Stern model of the electrical double layer

13

 

Small angle light scattering setup

[9

 

Determination of scattering angle

22

 

Single scattering limit for 0.3 In polystyrene particles

27

 

; potential for polystyrene particles.

29

 

4' potential trend

32

 

Typical SALS scan during slow flocculation

35

 

Typical SALS scan during rapid flocculation

35

 

E fleet of extent of heat treatment on dispersion stability

37

 

39

 

Effect of Stabilization on polystyrene particles

41

 

Stability Ratios of Native and Stabilized Particles.

45

 

48

 

Labview data acquisition program

55

 

viii

:0)
7t
E:-

sass“

e
d.
8

JV-GQPKJ

NOMENCLATURE

Definition

radius of polystyrene sphere
Harnaker constant

electrolyte concentration

diameter of sphere

bulk diffusion coefficient

local diffusion coefficient

protonic charge

initial scattering intensity
Boltzmann constant

Avogadro’s number

initial number concentration of particles
center to center separation distance
temperature

dimensionless separation distance
stability ratio

valence of charge

Greek Symbols
Definition
Debye length
attachment efficiency
dielectric constant
total energy of interaction between non-identical particles
repulsive energy
total energy of interaction
energy of vdW attraction
viscosity
Debye-Hackel parameter
wavelength
surface charge density
Stern potential
Zeta potential

ix

grits

cm

erg

M

cm

cmzs"

cmzs'l
dimensionless
arbitrary units
erg/K

mole"l

mL’1

cm

K
dimensionless
dimensionless
dimensionless

2219..

cm
dimensionless
dimensionless
erg

erg

erg

erg

kg m'ls'l
cm’l
nm
C cm
statvolt

mV

-2

1. INTRODUCTION

Colloidal behavior governs the performance of many industrial products, including surface
coatings, adhesives, textiles, paints, and synthetic rubber. Colloids are defined as disperse
systems of particles with one linear dimension between 10 nm and 1 um. Over this size
range, gravitational sedimentation is negligible, and Brownian motion is the major
mechanism for diffusion. The stability of a colloidal dispersion is determined by the
interplay between London and van der Waals attractive forces on one hand, and electrostatic
and steric repulsive forces on the other. Therefore, the rate of flocculation, which is a
measure of the stability of the system, is a sensitive method of determining the forces acting

between particles.

The classical basis for colloid stability emerged in the 1940’s when Derjaguin and Landau
(Russian scientists) and Verwey and Overbeek (Dutch scientists), working independently,
calculated the electrostatic repulsion between two particles on the basis of the interactions
between the double layers (Dcrjaguin and Landau, 1941; Verwey and Overbeek, 1948). Their
combined work, collectively known as DLVO theory, states that the stability of a colloidal
dispersion depends on the sum of the electrostatic repulsive forces (due to the overlapping of
ionic atmospheres around the particles) and the attractive forces (represented by London and
van der Waals forces). The core of the theory is that attraction decays as the square of the

inverse power of the separation distance and is nearly independent of electrolyte

to

concentration, while the repulsive forces fall off exponentially over a range equal to the
Debye length and are, therefore, strongly dependent on electrolyte concentration (Overbeek,

1982).

1. 1 STABILITYRATIO

The stability of a colloidal dispersion is a function of the electrolyte concentration, and is
usually presented as a plot of the log of the stability ratio (log W) against the log of
electrolyte concentration (log c) (Figure H). The plot yields two linear segments
representing a slow flocculation (reaction-limited cluster aggregation) region and a rapid
flocculation (diffusion-limited cluster aggregation) region In the rapid flocculation regime,
there is no energy barrier between the particles because the sum of attractive and repulsive
forces is zero or negative. Therefore, every collision results in coagulation. In the slow
flocculation regime, an energy barrier exists due to the dominance of repulsive forces over
attractive ones and the rate of coagulation is retarded As a result, only a fraction (NW) of
collisions leads to coagulation (Hidalgo-Alvarez et al, 1996). The intersection of the two
linear segments of the stability ratio curve is known as the critical coagulation concentration

(ccc).

1.2 PROBLEM DEFINITION

Polystyrene latexes are often used in the study of fundamental colloidal phenomena because
they are spherical and nearly monodisperse, with well characterized surface functional
groups. A large body of evidence suggests, however, that polystyrene latexes do not behave

as the classical theories predict. For example, experimental flocculation stability ratio curves

log W

 

rapid flocculation regime

critical coagulation concentration

 

{

log c

Figure 1-1: Typical stability ratio curve

4

are significantly different from those calculated on the basis of DLVO theory (Ottewill and
Shaw, 1966); the ionic strength dependence of polystyrene latexes exhibits a maximum in
zeta potential not predicted by theory (Rosen and Saville 1991, Hidalgo-Alvarez et al, 1992,
Elimelech and O’Melia, 1990); and particle deposition rates in the presence of repulsive

forces are grossly under-predicted by DLVO theory (Elimelech and O’Melia, 1990).

There have been many discussions in the literature as to the nature of the factors responsible
for the deviations between theory and experiment. This study is designed to investigate two
of the hypotheses proposed: 1) that a hairy layer on the surface of the particle induces surface
roughness, causing the particle to not conform to the spherical shape on which the classical
theories are based; and 2) surface charge heterogeneities produce a particle that cannot be
characterized by a constant surface potential (Litton and Olson, 1994; Seeburgh and Berg,

1995). The two factors are discussed in more detail below.

1.2.1 MODIFICATION OF SURFACE MORPHOLOGY BY HEAT TREATMENT

The “hairy layer” hypothesis suggests that a layer of flexible polymer chains is present on the
surface of the polystyrene particle (Rosen and Saville 1990; Midmore and Hunter, 1988;
Chow and Takamura, 1988). Zimehl and Lagaly (1987) suggest that hairy particles are one
of several types produced during emulsion polymerization. The hairs extend intojhe bulk
solution because of electrostatic repulsion between the ionic groups which terminate the
hairs and charges anchored to the surface. The hypothesis is that when the particle is heated

above its glass transition temperature (98°C), it becomes amorphous, and the sticky, mobile

5

polymer chains can rearrange and collapse on the particle surface, resulting in a smoother

particle (Rosen and Saville 1990).

Based on evidence from light seattering and photon conelation spectroscopy, Seebergh and
Berg (1995) concluded that the thickness of the hairy layer on PS particles was between 47
nm thick and that heat treatment collapses this layer. In addition, they report that heat
treatment reduces the absolute electrophoretic mobility and attribute this to a loss in surface
charge density by hydrolysis of the sulfate during heat treating (Seebergh and Berg, 1995).
Lastly, they observed a change in the critical micelle concentration (ccc) upon heat treatment

which they attributed to the collapse of the hairy layer (Seebergh and Berg, 1995).

Ofoli (1994) also investigated this factor by heat treating polystyrene particles for 6 hours
prior to measuring the rate of flocculation He found that results for heat treated particles
were in better agreement with theoretical calculations than those for unheated dispersions.
This raises the question as to what effect extended heat treatment would have on the
agreement of experimental flocculation rates with DLVO theory. The current work

addresses this issue.

1.2.2 REDUCTION OF SURFACE CHARGE HETEROGENEITY BY ADDITION OF A
SURFACI‘ANT
The deposition rates of particles onto granular quartz beds is often studied using latex

particles. The attachment efficiency (or) of the particles onto the beds is inversely

proportional to the stability ratio (W). Investigators have found that experimentally observed

6
attachment rates are grossly underestimated by DLVO theory when calculations are done on
the basis that charge is uniformly distributed on particle surfaces (Elimelech and O’Melia,
1990; Gregory and Wishart, 1980). On the other hand, theoretical models that incorporate
surface charge heterogeneity yield results that are closer to experimentally observed

attachment efficiencies (Kihara et al. 1992).

Litton and Olsen (1994) investigated this factor by assuming that the surface of the colloidal
latex was characterized by uncharged regions as well as regions negatively charged with
sulfate ions. It is known that anionic surfactants “shield” the uncharged regions by
hydrophobic attachment of the surfactant alkyl chain to polystyrene oligomer exposed at the
surface (Kayes, 1976; Kandori et al, 1989). To examine the effect of covering uncharged
regions of the latex with anions, Litton and Olsen (1994) added sodium dodecyl sulfate
(SDS) at slightly below the critical micelle concentration to their carboxyl latex dispersions.
They found that this addition greatly improved the agreement between experiment and

theory.

2.GENERALBACKGROUND

2.1 DL V0 THEORY

The quantitative theory which describes the interplay between electrostatic repulsion and van
der Waals attraction was formulated by Derjaguin and Landau (1941) and Verway and
Overbeek (1948), and is known collectively as DLVO theory. It states that the total energy of

interaction between particles ((1),) is the sum of the electrostatic repulsion ((DR ) and van der

Waals attraction ((DW. ). The repulsive forces are given by (Russel et al., 1989):

 

 

7k T)2 «a
=32 a[ 1
(DR nc ze e ( )
where
e'Q—l ve‘I’
a—+————- and = " 2
7 e"2+l (p H ( )

and, e is the dielectric constant, 2 is the valence of the counter-ion and e, is the protonic

charge.

The Debye-Hilckel parameter, 1: is given by:

z 87rcNA :2 e3
ckT

 

 

(3)

where c is the concentration of the electrolyte, NA is Avogadro’s number and e, is the

protonic charge.

8

The van der Waals attraction is calculated from (Pailthorpe and Russel 1982):

A 202 2a2 402)
(D = — e” + +ln‘ l—— 4
"N 6 [r2 ~4a2 r2 r2 ( )

where a is the radius of the sphere, r is the center to center separation distance, and Aefl is

 

the effective Harnaker constant from Lifshitz theory (Pailthorpe and Russel 1982)

The absolute stability ratio is the quotient of the rate of flocculation when interactive forces
between particles are absent, and the rate at any other electrolyte concentration:

rate when (I), = 0
rate when (I), at 0

 

Waba(ce) = (5)

where c, is the electrolyte concentration. The numerator of this equation represents the case
of rapid flocculation where every collision results in particles sticking together

(Smoluchowski kinetics); it is most nearly realized at high electrolyte concentrations.

2.2 THE EXPERIMENTAL S TABILITYRAIIO

An experimental absolute stability ratio can be directly calculated from small angle light

scattering (SALS) data using the following equation (Young 1991 ):

-l
2 1d] 377
W:-— -——— d =
7(10 d!) an I 4ltTno (6)

 

where 1,, and dlx'dt are the intercept and slope, respectively, of the scattering intensity versus

time profile, 17 is the viscosity of the dispersion, k is the Boltzmann constant, T is the

temperature and no is the initial concentration of particles (singlets).

2.3 THE THEORETICAL STABILITY RATIO

2.3.1.1 HOMO-FLOCCULATION
F uch’s equation for homo-flocculation can be used to ealculate the theoretical stability ratio:

W: 2a2:D——D(:lex¢)p[

where u is the dimensionless center-to-center separation distance between the particles at

I:(___:_):ldr (7)

their closest point of contact, and D,” and D(u) are the bulk and local diffusion coefficients,

respectively. The ratio of diffusion coefficients in the above equation represents a
hydrodynamic correction proposed by Spielman (1970) to account for the viscous drainage of
solvent from between particles as they approach one another:

D _6u +l3u+2

d3

 

 

D(u) 6u2 + 4n (8)
where u = r .. 20
a

2.3. 1.2 HETEROFLOCCULATION

The equations for heteroflocculation allows one to account for variations in properties such
as zeta potential and particle size. In the limit of monodisperse particles with no deviation in
zeta potential, the equations describing heteroflocculation reduce to Fuch’s equation given
above. Ofoli (1994) showed that the variation in zeta potential is important in the calculation
of the theoretical absolute stability ratio. As a result, it has been incorporated into the

theoretical calculations in this study.

lO

Plieve and Lin (1982) derived the following equation to account for variations in particle size

and surface potential:

e. exp(<I>y.(h)/kT)
: Rtio (herbal/0f"

 

(9)
where (I)g (h) is the total energy of interaction between particles i and j, h + R9- is the center-
to-center separation distance, and Rg‘ = a,- + a,- For the case of a distribution in surface

potential, the following equation for the mean stability ratio can be used (Prieve and Lin

1982y

1‘”pr )pcr )a‘l’id‘l’

<W >= l” l Pinfld‘i’zfllawaw (10)

 

where <WW> is the mean stability ratio due to a distribution in surface potential, p(‘I’,) is the
probability density for Stem potential ‘1’, and WOPI, T2) is the heterogeneous stability ratio

defined in Eq. (9).

2.4 SMALL ANGLE LIGHT SCA IT'ERING

Light scattering is well suited to measuring the rate of flocculation because of its high
sensitivity to small changes in particle size. Lord Rayleigh (1918) laid the foundation for the
theory of light scattering in the early 20"“ century by applying the electromagnetic theory of
light to small, non-absorbing particles in a gaseous medium. He showed that if a particle is
smaller than 1/20th of the wavelength of the incident radiation, it will scatter light in
proportion to the square of its volume. To extend the applicability of Rayleigh scattering,

Debye and Gans introduced a correction to this theory by incorporating a form factor that

ll

accounts for interparticle scattering from different volume elements within a particle larger
than the Rayleigh limit (Kerker, I969; Oster and Riley, 1952). Debye also argued that, at
small angles, the Rayleigh limit can be further relaxed because interference effects in the
forward scattering direction become negligible. Recently, Ofoli (1994) produced
experimental evidence that, at small angles (2° or less), the regime of Rayleigh scattering can
be extended to D 5 0.95 um for probing polystyrene spheres in water with a helium-neon

(HeNe) laser.

The small angle scattering of light fiom a dispersion of identical primary particles with time-

varying floc sizes is (Lips and Willis, 1973; Zeichner and Showalter, 1979):

1(9.t)=in,(t)izl.(9) (M)

i=1
where n 1(1) is the number of flocs at time t containing j primary particles. For a flocculating

dispersion of primary particles, the rate of disappearance of singlets from the system

-dn, / dt , based on Eq. (11) is (Lips and Willis, 1973; Zeichner and Showalter, 1979):
- 2”). = 10. w) (12)
dt 1, dt

where n0 is the initial number of singlets and I0 is the intensity of the incident radiation (Lips
and Willis, 1973; Zeichner and Showalter, 1979). The experimental flocculation rate can be

calculated by substituting the initial intensity and slope 51%, of the intensity vs. time plot

for the flocculating dispersion into Eq. (12).

l 2
2.5 ELECTROPHORETIC MOBILITY

The repulsive forces between colloidal particles are described by a model of the electrical
double layer (EDL). The theory of the EDL deals with the distribution of counter-ions and
co-ions at the surface. Helmholtz (1879) first proposed a model of a fixed layer of counter-
ions adsorbed to the surface of the colloid This was modified by the Gouy—Chapman model
which assumed that the electrical properties at the surface of a colloidal particle are a balance
between electrical forces which tend to attract counter-ions, and thermal motion which tends
towards a uniform distribution of the ions (Gouy, 1910; Chapman, 1913). The model
predicts that the combined effect of this competition is to produce a “diffuse” electrical

double layer rather than a fixed layer of ions at the surface.

The two schools of thought were combined by Stern (1924) who proposed that the electrical
double layer is a combination of adsorbed and diffusing ions. Stern introduced a correction
for the finite size of the ions in the first layer adjacent to the charged surface, and argued that
electrostatic and van der Waals forces near the surface might be enough to overcome the
thermal motion of the ions in the vicinity of the surface. The EDL in Stem’s model is,
therefore, divided into two parts: a compact layer of attached counter ions at the surface,
surrounded by a diffuse collection of co- and counter-ions extending from the particle surface
into the bulk fluid (Figure 2-1). The “thickness” of the diffuse layer is given by the Debye

length (l/rc) (Laidler and Meiser, 1982):

K"=\/ 8kT (13)

'3 1
87! c NAz‘e“
0

 

 

13

where c is the concentration of the electrolyte, NA is Avogadro’s nmnber and e, is the
protonic charge. It can be seen from Eq. (13) that the Debye length decreases as the

electrolyte concentration increases, leading to a decrease in the diffuse double layer.

Diffuse Layer

 

+ ' :
. . \y. - Surface Potenml'
Stern Plane "' i g i i we- Stan Potential

i; -Zeta Potential at shearplam
I/K- chyclcngth

 

Figure 2-1: Stern model of the electrical double layer

Ifan electric field is applied to an aqueous colloidal dispersion, a force is created in both
parts of the double layer. The charged surface of the colloid and the solvent inside the shear
plane tend to move in the attractive direction, while the ions outside the surface of shear
move in the opposite direction. The movement of the colloidal particle in response to this

applied potential gradient is the electrophoretic mobility.

3. OBJECTIVES

DLVO theory, generally accepted as the preeminent theoretical model for particle-particle
interactions, has been shown by many researchers to predict a steeper stability ratio curve
than is obtained experimentally during the slow flocculation of colloidal particles. Two of
the more likely reasons for this classical discrepancy are that 1) due to surface roughness,
particles do not conform to the smooth, perfectly spherical shape assumed by the classical
theories; and 2) particles are subject to heterogeneities in surface potential and cannot be

characterized by a single constant potential.

While both of these factors are difficult to account for directly in theoretical calculations,
experimental techniques are available which enable one to evaluate their effects. For
example, in a recent study, Ofoli and Prieve (1997) annealed polystyrene particle surfaces by
heat treating for six hours to reduce the degree of surface roughness, and showed that this
resulted in a marked reduction in the discrepancy between theory and experiment. One of
the goals of the current study was to extend the work of Ofoli and Prieve (1997) to evaluate
how the extent of heat treatment affects the discrepancy between theory and experiment

during slow flocculation of colloidal species.

In another study, Litton and Olson (1995) showed that surface charge heterogeneities can be
reduced by adding a surfactant to a dispersion at a concentration below the critical micelle

concentration (cmc). They reported that experimental attachment efficiencies of carboxyl

14

15

latexes onto granular quartz sand in the presence of 10'3M sodium dodecyl sulfate were
closer to the theoretically derived values. The second goal of this study was to investigate the
effect of adding a surfactant to the colloidal dispersions prior to measuring the initial rate of

flocculation.

Specifically, the objectives of the study are:

1. To evaluate the effect of the extent of heat treatment beyond six hours on the agreement
between experimental flocculation rates and those calculated on the basis of DLVO
theory; and

2. To assess the effect of adding sodium dodecyl sulfate (SDS) to mask surface charge

heterogeneities on this classical discrepancy

4. MATERIALS AND METHODS

4. 1 EXPERIMENTAL DESIGN

Clean, surfactant-free polystyrene particles were purchased from lnterfacial Dynamics
Corporation (IDC, Portland, OR). Table 4-1 lists the experiments performed to evaluate the
effects of heat treatment and stabilization with a surfactant ElectIOphoretic mobility
measurements could not be completed on all samples due to time constraints. The 24 hour
samples were chosen for electrophoresis measurements because they would provide the

largest potential beneficial result of surface annealment.

4.2 SAMPLE PREPARATION

4.2.1 HEAT TREATMENT or COLLOIDAL DISPERSIONS
The dispersion, originally at a number concentration of 5.9x1012 particles/mL, was diluted to

make a 3.34x10lo particles/ml stock solution (subsequently referred to as “stock colloi ”).
The concentration of the stock colloid (0.05 vol%) was chosen based on evidence presented
by Rosen and Saville (1991) that suspensions above 0.5 vol% tended to flocculate during
heating. The samples to be heat treated were taken directly from this stock and sealed in a
Parr bomb (Parr Bomb Corp., Moline, IL). The bomb was placed in an oven at 120°C and

heated for 6, 12 or 24 hours. The heated dispersions were slowly cooled,

16

17

Table 4—1: Experimental Design

 

Designation

Treatment

Light scatteriflngr

Electrophoresis

 

Native

none, diluted directly from
IDC stock

yes

yes

 

6 hour heat treated

native particles were
heated for 6 hours

yes

 

12 hour heat treated

native particles were
heated for 12 hours

yes

 

24 hour heat treated

native particles were
heated for 24 hours

yes

yes

 

Stabilized

native dispersion was
stabilized with SDS below
the cmc

yes

yes

 

6 hour heat treated

native particles were
heated for 6 hours, then
stabilized with SDS below
the cmc

yes

[10

 

12hOIIIheattreated

native particles were
heated for 12 hours, then
stabilized with SDS below
the cmc

st

no

 

 

24 hour heat treated

 

native particles were
heated for 24 hours, then
stabilized with SDS below
the cmc

 

yes

 

yes

 

 

18

with the oven temperatures successively stepped down to 90°C for four hours, 60°C for four
hours, and room temperature for four hours. Since there was no sedimentation at the bottom
of the sample vial after heat treatment, it was assumed that flocculation did not occur during
the process. The dispersions for light scattering and electrophoresis were prepared directly

fiom these samples.

4.2.2 STABILIZATION or DISPERSIONS WITH SODIUM DODECYL SULFATE (SDS)
A stock solution was prepared from solid sodium dodecyl sulfate (Boehringer Mannheim

Laboratory Reagents, electrophoretic grade) by adding the appropriate weight of dry solid to
a clean, dry volumetric flask. Double filtered distilled water was then added. The resulting
solution was sonicated for 30 minutes to insure complete mixing. After cooling, the solution
was brought to volume. An aliquot was taken from this stock solution to bring the final SDS
concentration in the stock colloid solutions to a concentration of 1x10'5 M. This
concentration was used to insure that micelles did not form (the cmc of SDS is 1x10’3 M at
30°C). The heat treated samples were stabilized only after the heat treatment procedure

described above.

4.3 SMALL ANGLE LIGHT SCA TTERING

4.3.1 LIGHT SCATTERING SETUP
Small angle light scattering (SALS) was used to measure the absolute flocculation rate, using

the setup described by Young and Prieve (1991 (Figure 4-1). In this setup, light from a 1

mW Helium-Neon (HeNe) laser (A =633 nm) passes through a 10x beam expander (2) to

 

 

 

 

 

ll

\

b)
A
Lil

 

 

 

 

 

 

 

 

 

 

 

 

Figure 4-1: Small angle light scattering setup

20

increase the diameter of the beam from 0.6 to 6.0 mm, a neutral density filter (3) to reduce
the incident intensity, a 3mm aperture plate (4) to eliminate all but the peak intensity of the
Gaussian distributed incident light, and into a 1 cm cuvette containing the sample (5). To
insure that only the light scattered at small angles is detectable, two 270° annular slit plates in
series (6 and 7) configured to produce an angle of 2° were placed in the path of the scattered
light. The direct incident light was attenuated at the first slit plate with a Rayleigh horn

mounted in its center.

The light scattered at an angle of 2° from the horizontal passes through a plano-convex lens
(8) which focuses the beam on the detector (10). A 10 nm bandpass filter placed in front of
the detector (9) assures that the only wavelength reaching the detector is 633 :t 10 nm.
Component 11 is a light tight black box that surrounds the assembly so that no stray light
reaches the detector. The entire assembly is mounted on an Oriel 2-m optical rail (Oriel
Corporation, Stratford, CT) and housed in a NuAire clean air laminar flow hood (NuAire,
Plymouth, MN ) to alleviate dust contamination. The output of the photomultiplier tube
detector is monitored by an Oriel Model 7070 combination ammeter and high voltage power
supply (Oriel Corporation, Stratford, CT) which is connected to a National Instruments data
acquisition board (AT MIO 16XE-50) using a standard PC. A LabView (Version 3.1,
National Instruments, Austin, TX) data acquisition software was used to collect the intensity

measurements.

21
The configuration of the annular slits to produce the scattering angle of 2" is shown in Figure
4-2. The angle is defined by 0 in the figure and is obtained by taking the average of the
angle formed by a line extending from the bottom of the first slit to the top of the second slit,
and the angle formed by a line running from the top of the first slit to the bottom of the

second slit. In our set-up, the scattering angle is 16° +/- 0.08.

4.3.2 NECESSITY or CLEAN EXPERIMENTS
Since the intensity of light scattered is proportional to the square of the volume of a particle,

dust, dirt, crystals or entrapped air bubbles will scatter much more light than the colloidal
particles. A speck of dust, for example, is on the order of 1 um in diameter while the
particles are 0.3 pm. In relative terms, a speck of dust will scatter more than one thousand
times as much light as each of the particles. Consequently, many precautions were taken to

insure that no foreign particles were incorporated into the samples.

All work, including mixing of dispersions, was performed in a NuAire laminar flow hood
Aliquots of the samples were measured using an Eppendorf pipette (Brinkmann, Westbury,
NY) fitted with sterile disposable pipette tips. The tips were first blown free of contaminants
using compressed air (Aero-Duster, Miller Stephenson, Danbury, CT). All water used in the
preparation of samples or cleaning of glassware was double filtered through a 0.02 pm filter
prior to use. Glassware was washed with filtered water, then shaken dry inside the laminar
flow hood Salt solutions were filtered after preparation and again before use. In addition,
all data sets were examined for intensity spikes, generally an indication of foreign matter

contamination.

22

 

 

 

 

 

 

Cuvette
Incident Scattering
Laser Beam Angle, 9
Scattering Vol.

 

 

 

 

 

Annular Slit Plates _______,.

 

 

 

 

 

Figure 4-2: Determination of scattering angle

23

4.3.3 DETERMINATION OF SINGLE SCATTERING LIMIT
Using the definition proposed by Ofoli (1994), the single scattering limit was determined by

the following procedure. A cuvette was filled with two milliliters of double filtered distilled
water and placed in the sample holder of the SALS apparatus. The shutter of the
photomultiplier tube was opened and the scattering intensity was measured for a period of 2
minutes at 2 scans/second The average intensity over this time period was used for the

background reading.

An aliquot of the stock solution was then added to the blank solution to produce a colloidal
dispersion The scattering intensity of the dispersion was measured for 10 minutes at 2
scans/second. The background reading was subtracted from the average intensity of the
dispersion to obtain the scattering intensity at that number concentration (data point). These

data were collected in the absence of electrolytes.

It should be noted here that the ten minutes of scanning for the colloidal dispersion was
probably unnecessary. It was determined by repeated experiments that 2 minutes was
sufficient for a baseline reading in the absence of colloidal flocculation Ten minutes was

chosen only as a precautionary measure to make sure that enough data points were obtained.

4.3.4 INITIAL FLOCCULATION RATE EXPERIMENTS

Scattering intensity measurements were obtained by SALS over electrolyte concentrations

ranging from 0.0001 to 0.2M KCl for unstabilized samples and 0.01 to 0.0st KCl for

24

stabilized samples. Lower salt concentrations were used for the native particles because they
are inherently less stable; therefore, smaller amounts of electrolyte are required to achieve
flocculation The following procedure was used to obtain the measurements. A stock
electrolyte solution with no particles was prepared by adding potassium chloride (KCl)
(Fisher Scientific, 99% pure) to double filtered distilled water. An aliquot of this solution
was diluted to the appropriate electrolyte concentration, and 2 mL was placed in a cuvette.
Intensity measm‘ements were collected on the blank solution for 3 minutes at a rate of 2
scans per second Here again, the three minutes for a background measurement was only a

precautionary measure. The measurements were averaged to obtain a background reading

With the shutter on the photomultiplier tube closed, one milliliter of the colloidal dispersion
was added to the blank solution, and the data acquisition system started The cuvette was
gently tipped back and forth to assure complete mixing of the particles in the electrolyte
solution, taking care to not introduce air bubbles. After the colloid was added, the cuvette
was placed back into the sample holder, the shutter on the PMT was pulled open, and
intensity measurements were collected for 30 minutes at a rate of 2 scans per second The
average value obtained from the blank measurement was then subtracted from each of the

intensity readings to obtain the scattering profile of the flocculating dispersion.

A Laser Zee Model 501 (Pen Kern, Bedford Hills, NY) was used to measure the
electrophoretic mobility of the particles. It uses the Smoluchowski model to convert

electrophoretic mobility to zeta potential. The measurements were taken at the upper

25

stationary layer of the instrument. A standard colloid was measured before each set of runs,
to insure correct operation of the instrument. The sample aliquot was carefully added to the
electrophoresis cell with a syringe to avoid introducing air bubbles. The cell was then placed
on the Laser Zee and the zeta potential and conductivity were measured according to

instrument instructions. The temperature and pH were also recorded

Zeta potential data were obtained for four sets of treatments: native, 24 hour native heated,
stabilized, and 24 hour heat treated stabilized dispersions. For each treatment, mobility

measurements were made at five electrolyte concentrations as described earlier.

5. RESULTS AND DISCUSSION

5. 1 SINGLE SCATTERING LIMIT

To determine the single scattering limit, data were collected over a concentration range of
3x107 to 7x109 particles/mL, using SALS as described earlier. A regression line was
computed using the first five data points. Since the scattering intensity is proportional to the
incident radiation, the total scattering intensity must be linear with respect to particle
concentration when all particles see the same incident light with little or no attenuation of the

laser beam.

Based on this concept, the first deviation of the experimental scattering data from this line
was defined by Ofoli (1994) as the termination of the single scattering regime, and is
considered to mark the onset of multiple scattering (Figure S-l). For this data, the
relationship between particle concentration and scattering intensity remains quite linear until
a concentration of about 1.5x109 particles/ml, which is in agreement with the scattering
concentration limit established by Ofoli (1994) for this particle size. Based on this limit, a
concentration of 8x108 particles/mL was chosen for all flocculation experiments, to insure

that the single scattering limit is not exceeded.

26

27

 

O.”

0.005 4»

Scattered Intensity, AU
.0 .o
.3 §

E

0&1 <

 

A

_A

o scattering intensity
—-—trendline

 

 

0.00900

1.00909

v7

2.(IE+09 3W“ 4&1009 SKEW 6.02m

particle concentration (parficlelmL)

Figure 5-1: Single scattering limit for 0.3 pm polystyrene particles

7.00909

28

5.2 ELECTROPHORESIS

Electrolyte solutions were prepared to give a final concentration spanning a range of about
0.01 M KCl to 0.25 M KCl. The 25 mL aliquot necessary for electrophoresis experiments
was prepared by mixing 23 mL of the stock solution with 2 mL of electrolyte solution The
stock solutions were sonicated for 15 seconds to break up any flocs that might have formed
during refiigeration. The concentration of electrolytes in the final solution was determined
by plotting a stande curve of conductivity versus concentration for five electrolyte

solutions. The standard curve was then used to calculate the electrolyte concentrations.

5.3 ELEcmOPHOREnc MOBILITY

Zeta potential data were obtained for four of the treatments defined earlier: native, 24 hour
heat treated native, stabilized, and 24 hour heat treated stabilized dispersions. A cubic spline
was fit to the five data points (Figure 5-2) to obtain an expression to be used in the
calculation of the theoretical stability ratios. The zeta potentials decrease monotonically with
increasing electrolyte concentration, as predicted by the Gouy-Chapman equation, which

applies in the limit as RU —-> oo (Hunter, 1981):

. e6
0' - AJEsrnh 2kT (14)

Assuming that the surface charge (0') is constant, it is obvious fiem Eq. (14) that the 4'
potential must decrease as the electrolyte concentration (0) increases. The trend in the
mobility data is consistent with that reported by other researchers (Seebergh and Berg, 1995;

Ofoli, 1994).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

50,,. 'T 5011’ ’ ..
1 _ 45 1 c Stabilized 1_
‘5 1" ° Nam Cubic Spline
40 1., \ —— Cubic Spline 1 > 40 3» , l
> i i" 35 4. I
’2': 30 gr .\ g 1 .\
E 25 i i ‘ c 25 1b \\ if
° 1 “e 3 “\\ E
‘5 20 :i "\ 02° ‘* k, ‘t
a “ \\ Q15 ‘ \\_ :
U 151 ‘~2._\. U 4:” \‘L;
10 5+ 10 1* 'r
5 ,1 5 i' i
01 fi-f: .. .rvfl +. 1*, 01...;jsnrnv‘1. nava
0 0.05 0.1 0.15 0.2 0 0.05 0.1 0.15 02 025
electrolyte concentration, M electrolyte concentration, M
(a) Native (b) Stabilized
50 , ~ ~ -— so ..
45 ¢\ 5 24 hr. ht Native ,5 . 24 hr. ht. Stab.
40 \ - Cubic Spline ,0 -—--~ Cubic Spline
> i \\ i > “x
E35 i E 35 \‘x
J30 J
.3 \ a! w \\
§25 ‘5 25 "x
O 20 \H— _, g 20 K“ ‘ J“\\.
Q15 N \““‘“‘~~ 0. ‘19
U. U 15
10 g 10
5 g 5
0 . . HM-.. ....... ..
0 0.05 0.1 0.15 0.2 0 0.1 0.2 0.3
electrolyte concentration, M electrolyte concentration, M
(c) 24 hour heat treated native (d) 24 hour heat treated stabilized

Figure 5-2: C potential for polystyrene particles.

 

 

30

The Lazer Zee only gives an average zeta potential, and is not capable of resolving the
distribution of zeta potentials for the dispersion. Since this distribution is essential in our
theoretical calculations, we have assumed an average standard deviation in the zeta potential
of 12% for native unstabilized particles, and a 15% for heat treated particles, based on data
from Ofoli (1994). In the absence of additional data, the same values were used in the
theoretical calculations for stabilized particles, although Ofoli (1994) did not use these in his

work.

A maximum was observed in the electrophoretic mobility profiles at very low electrolyte
concentrations for the native and the native heat treated particles. There were no maxima in
the profiles of either of the stabilized particles. The fact that a maximum was not observed
for the stabilized particles was most likely due to the fact that readings were not taken at a
low enough electrolyte concentration. The observed maxima occurred well below the range
of electrolyte concentrations required for the theoretical stability ratio calculations in this

study, therefore, it will not be a factor in any comparisons between theory and experiment.

The C potentials for the four cases studied are presented in Figure 5-3. Stabilization increases
the C; potential (Figure 5-3 a and b), which is to be expected, and is consistent with the
assumption that the hydrophobic tail of the surfactant adsorbs onto uncharged regions of the
colloid (Kayes, 1976; Kandori, et. al, 1989). The hydr0philic head of the surfactant would
then extend from the surface, adding steric as well as electrostatic stabilization, thus

increasing the magnitude of the Q potential. These observations are consistent with other

31

electrokinetic studies of anionic surfactants added to polymer latex particles (Litton and

Olsen, 1994; Kandori, et al, 1989; and Kayes, 1976).

Heat treating the colloidal dispersions results in a lowering of the C potential, as can be seen
in Figure 5-3 c and d. This is consistent with what has been reported by other researchers
(Rosen and Saville, 1990; Seebergh and Berg, 1995; Elimelech and O’Melia, 1990).
Seebergh and Berg (1995) have shown that the surface charge density on polystyrene
particles decreases upon heat treatment, most likely due to hydrolysis of the sulfate groups on

the polystyrene particle surface by the following reaction:

0
ll
R-O-S—O'M+ 5’3» R-OH + M’HSO,‘
H h“
0

They reported that half of the sulfate functional groups were converted to uncharged

hydroxyl groups after 12 hours of heat treatment.

The reason for the crossover in the zeta potential profiles of native and heat treated particles
Figure 5-3 c and d is not clear. Fortunately, the crossover points occur well below the
theoretical critical coagulation concentration (coo), as will be shown later. Therefore, they
do not affect the comparison of theoretical and experimental stability ratios in the slow

flocculation regime.

 

 

 

 

 

 

l”
i 50 .-. .
i 45 — ' Native
§ ,0_ \ .,_ Stabilized
. \ \.\
i =7 so «
a g 25" .
l \ ‘\._
g % zo - \
m \> _\ .
l U 15- g- 1 .1
i 10»
i 0 i J. ‘
l o 0.1 0.2 03
5L electrolyte concentration, M
(a) Native and Stabilized
50 1- .
45 1. \ -1—~- Native
40 ..
> \h 24 hr. ht. treated
E 351' \‘g. native
: 25 - .\
§ 20» \2 a.
0- \:<::;T‘\_
U 15 - x-..
10 »
5 -.
0 l l l 1
0 0.05 0.1 0.15 0.2

electrolyte concentration, M

 

(c) native and
24 hour heat treated native

 

 

32

 

 

5° ..
45 ‘ 7, ,, —- 24 hr. ht. treated ‘
‘ 40 native -
. > \ ————— i 24 hr. httreated 1
i E 357 \1 ‘ stabilized
l - ‘ \
: E 30 “' it \\|\
l E 25 1— ‘
a w '
'6 2° 3 XRILT'“ ‘\
n 15 «- \\"
3 U
'2 10 it
‘ 5 i.
l 00 I

i electrolyte concentration, M

 

(b) 24 hour heat treated native and 24 hour

 

 

 

 

 

heat treated stabilized
l
f 50;-~~~~- - —
3 is i. —— Stabilized
> ‘° \x ——24br.httieatedf
l E 35. stabilized
. —‘ 30‘ \¥.\.
.‘2 \
i g 25 \I'\\:\_‘
*5 20. ‘-\-_x:\\
a} 151 ‘\
10 ‘7
j 5
l o f 4
0 0.1 0.2 0.3
electrolyte concentration, M
(d) stabilized and
24 hour heat treated stabilized

Figure 5-3: C potential trend

 

33

5.4 CRIUCAL COA GULA norv CONCENTRATION

The experimental critical coagulation concentration (ccc) for the native particles was f0lmd
to be 0.15M KCl, which is in agreement with results reported by other researchers (see Ofoli
1994, for example). No data were collected beyond an electrolyte concentration of 0.2 M
KCl for the unstabilized surface annealed particles because it was not apparent that this value
would be less than the ccc. Since the focus of this research is on the slow flocculation

regime, not having data beyond the ccc should not present any particular difficulties.

The critical coagulation concentration for the stabilized particles, both native and heat
treated, was calculated as 0.4M KCl. The increase in ccc relative to the value for the native
particles appears to be a logical trend Assuming the surfactant alkyl chains adhere to the
polymer with the ionic head sticking out into solution, this would add electrostatic stability to
the dispersion. Therefore, a higher concentration of electrolyte would be required to firlly

depress the electrical double layer.

5.5 ma STABILITYRAHO

Figure 5-4 and Figure 5-5 are typical scattering intensity profiles for slow and rapid
flocculation, respectively. The sharp rise in intensity after one minute is due to the fact that
data collection was initiated immediately upon adding the colloid to the electrolyte solution
as described earlier; however, the shutter to the photomultiplier tube was opened only afier
the dispersion had been completely mixed and the cuvette had been placed into the sample

holder. The $10pe (dl/dt) and intercept (IQ) of each scan were obtained from a regression of

34

the initial part of the data. Using these values, the experimental stability ratio was calculated

from Eq. (6).

The experimental stability ratios were lower than those reported in the literature. Ofoli
(1994), for example reported a value of about 1000 for 0.301 um polystyrene latexes at
electrolyte concentrations of 0.01M NaCl. Additionally, Kihara (1994) reported a stability
ratio of about 1000 at 0.001M KNO; for 0.26 um polystyrene particles, which is also
significantly higher than the ratio measured in this work. These differences are most likely
due to problems with establishing a unique slope for scans at very low salt concentrations, as

discussed later.

The theoretical stability ratios were calculated using Eqs. (9) and (10), incorporating the zeta
potentials detemrined from electrophoretic mobility measurements, and the assumed
standard deviation, using a MathCad program developed by Ofoli (1994). The theoretical
stability ratio shows a much higher sensitivity to electrolyte concentration than the
experimental data, which is consistent with the trends reported in other flocculation and

deposition experiments (see Ofoli 1994, for example).

 

 

scattering intensity

35

 

 

 

 

 

 

     
     

 

 

16 ~
14
12
10 .
8 _ y = 1.17E—04x + 8.64E—01 ‘
6 i
4 ;
2 - -1- 2 2 2 ._S A 24- E
0 ~——-——‘-——' ww' fir——— T 7 7 , ~l
'i (l 500 1000 1500 20:00
-6 t
time (s)
Figure 5-4: Typical SALS scan during slow flocculation
20 ~-
15 <-
y = 6.591E-03x + 7.820E-01
10 4»
5 ~—
0 i z 1 fi'
(J 500 1000 1 500 2000
time (s)

 

Figure 5-5: Typical SALS scan during rapid flocculation

 

5.5.1 Earner OF THE EXTENT OF HEAT TREATMENT

These results have confirmed that the overall effect of heat treatment is to reduce the stability
of the given colloidal dispersion It is generally accepted that heat treatment also anneals the
surface, making the particle smoother by collapsing any polymeric hairs that may have
extended into the solution It has also been shown that the surface charge density on
polystyrene particles decreases upon heat treatment, most likely due to hydrolysis of the

sulfate groups on the polystyrene particle surface.

To demonstrate the effect of the extent of heat treatment on both experimental and
theoretical calculations, the stability ratios at the lowest electrolyte concentration measured
are plotted as a function of the duration of heat treatment for both native and heat treated

particles (Figure 5-6). The percent change in stability ratio with respect to heating time is

also tabulated in Table 5-1.

Table 5-1: Relative reduction in stability ratio upon heating, %

 

 

 

[ 6 hours of heating 12 hours of heating 24 hours of heating 1
[ native 89 -2 2 |
L stabilized 70 39 76 I

 

 

 

 

The change in the stability ratio is very substantial after 6 hours of heating for both the native
and stabilized particles. After 6 hours, the stability ratio of the native particles shows no

further reduction in stability, while the stability ratio of the stabilized particles shows only

limited further loss in stability.

Shhlllty ratio. W

37

 

   
  

 

+ name particles in 0.0005 M KCl

 

+ stabilized particles in 0.01 M KCl

 

 

 

 

0 5 10 15 20
heat treatment (hoard

Figure 5-6: Effect of extent of heat treatment on dispersion stability

38

Clearly, there is a competition between the destabilization effect caused by heat treatment
and the re-stabilizing influence of adding surfactants to the dispersion. For native particles,
heat treatment destabilizes the particles completely after 6 hours, and further heating has
little or no effect on reducing the stability ratio. When SDS is added after heating, some
degree of stability is regained by the dispersions. But even adding SDS has its limit, because

after 24 hours of heating, the stability ratio is the same for both the native and stabilized

particles.

The same information can be seen in a different way by examining Figure 5-7. Here both the
experimental and theoretical stability ratios have been plotted for comparison The arrows
are meant to aid in comparison of stability ratios at a given electrolyte concentration A
decrease in stability is observed for the native particles in both the experimental (Figure S-7a)
and theoretical (Figure 5-7c) curves. It appears that the decrease in the theoretical stability
ratio at a given point with heat treatment is primarily due to a decrease in the ccc of the
dispersion (from 0.82 to 0.057). The decrease is most likely attributable to a reduction of the
steric barrier to coagulation following surface annealment. The fact that this is not
observable in our experimental study is probably because enough data points were not

collected around the 000, as explained earlier.

The same decrease in stability with heat treatment is observed in the both the experimental

(Figure 5-7b) and theoretical (Figure 5-7d) stability ratios for stabilized particles. The

 

39

 

 

 

 

 

 
 
   

 

 

 

 

 

 

 

 

 

 

electrolyte cooncentration, M

 

 

    
   

 

 

0.1 1
electrolyte concentration, M

_. ....1goo..:_ -- .-.. 2.-.... _ ...-.-1000 1-
. 3 uStabilized :
1 o 6 hour heat treated
9 1 A 12 hour heat heated
1‘: ‘ o 24 how heat treated
{ .1 ° ‘._ 100 3a 100 1.
i e ::‘:... ° -- j 3
°‘. 1 , :
'3' ‘ 2',- \ 10 31 10 3
I mstabilized Kg, : :
. 6 hr. ht. treated I.
o 12 hr. ht. treated . _ o 8
a 24 hr. ht. treated '.I . 4
0.0001 0.001 0.01 0.1 1 0.01 0.1 1 . 10
electrolyte concentratlon, M electrolyte concentration, M
(a) Experimental Native (b) Experimental Stabilized
, - -~ 11400011 ~—- .. ,— . “100011
3 Native 1 1
E l '-. ‘
1 1 .--_--. 24111.01 1 r
t 9 treated
1 2 n -- :l' 100 1*
'1 I 1 i
3
10 31 ———Stabized, 10 3-
i. i
1 --- --- 24 hrJlt. ‘\ 1
< treated x.. t
L , Stabilized A
0. 01 0.01

 

(0) Theoretical Native

 

(d) theoretical Stabilized

Figure 5-7: Effect of Heat Treatment on Polystyrene particles

 

 

40

change in stability ratio is not as drastic as that predicted for the native heat treated particles,
however (from 0.096 to 0.089). Once again this is probably due to the competition between
heat treatment and stabilization. Once the particles are heat treated, it is assumed that all of
the polymeric hairs that contributed to the steric stabilization have collapsed to the surface.
The addition of the surfactant, however, reintroduces some steric interactions because of the
hydrophillic head groups sticking out from the surface. Thus, the overall effect of the heat

treatment is reduced, as is reflected in the ccc.

5.5.2 EFFECT or STABILIZATION wrrrr SODIUM DODECYL SULFATE
The consensus of various researchers is that the addition of a surfactant stabilizes a particle

by the adsorption of the hydrophobic tail onto the surface of the colloid, adding both steric
and electrostatic stabilization to the dispersion To examine the effects of adding a surfactant
to a colloidal dispersion, the stability ratio of the stabilized particles are compared to their
native counterparts in Figure 5-8. Again, the arrows are meant to aid in the comparison of

stability ratios at a given electrolyte concentration.

A similar increase in stability upon addition of the surfactant was observed for both the
experimental (Figure 5-8 a) and theoretical (Figure 5-8 0) curves. The slight increase in ccc
that is apparent in the theoretical stability ratio (from 0.082 to 0.096 M KCl) would not be
easy to detect experimentally, particularly in light of not having enough data points in the

immediate vicinity of the ccc.

 

 

 

4l

 

 

 

 

 

 

 

electrolyte concentration, M

 

 

 

' 01" ‘ \\ 1000.0 2» 1000
3_ "_ : l j 15 Experimental- 24 hr ht stab.
9. ‘ ' r Experimental - 24 hr. ht. native
1 '\ \. ' r
. -\
‘.\ 100° ’7 A, 100
\ A\ < ' \ o
‘ \ . \ A l \
r . ‘1 l ‘\ ‘ 0 \
a1 ‘10 y - g 3 . \t; .
f \ll‘ , I ‘\ L w
. \\ 110.01: \I‘ \K 10.
I .\ \'\ E L\\ .
mama ‘xlt : .
' . . j. l .O 9
; a Stabilized 3! .' . 1'
W’ l V' r I if" ‘I 1
0.0001 0. 001 0. 01 0.1 1 0.001001 1
electrolyte concentration, M 1 electrolyte concentration, M
_.,.._fl#_.. fl. .- __ *__ _ __ _ - ..fi. ._. .2-..__.V ._ ___. .;‘__fi__ .-__.J
(a) Experimental native (b) Experimental heat treated
f l . ------- Native 1 3 ‘3 ------- 24 hr. ht
w l ......
:E"; Stabil' ed ', I ; Native
-.;_ " 'z :f; ~24hr.ht°o_
1 1w " l . 1 ‘1' t
.. 1 i ; l 1 mated
1; i g; i. Stabilized
‘ 10 .. E 1 '1 10 .-
o. 01 1 0.01 0.1 1

 

1 electrolyte concentration, M

 

(c) Theoretical native

(d) Theoretical heat treated

Figure 5-8: Effect of Stabilimtion on polystyrene particles

42

The same increase in stability upon addition of surfactant is observed in both the native
(Figure 5-8 b) and theoretical (Figure 5-8 d) heat treated particles as well. The change in
stability of the theoretical stability ratio of the particles is much more drastic than that seen
for the experimental ones. The increase is apparently primarily due to a change in the ccc
for the particles (from 0.57 to 0.89 M KCl). That this is the largest change in ccc seen for
any of the dispersions is explained by thinking about the hypothesized effects of each
treatment. The heat annealment of the particles is expected to collapse the hairy layer of the
particle thereby reducing the ccc. On the other hand, the stabilization of the particles is
expected to add stability to the dispersion because of the steric and electrostatic interactions
between stabilized particles. The combination of these two effects should result in the

largest difl‘erence in ccc, which is observed in the theoretical measurements.

5. 6 DIFFICULTYIN CHARACTERIZHVG SLOPES 0F SCA TTERING INTENSITY

PROFILES

It is difficult to obtain a unique value for the stability ratio at small electrolyte concentrations
because the scattering intensity scans are very flat It is obvious from the scan of the
scattering intensity with time measurements that there is some increase in intensity with
respect to time. Determining exactly what this increase is, however, is complicated by the
fact that the slopes are very small. Additiondly, due to the nature of the samples and the

detection system used, there is always some scatter and drift in the data.

43

Take, for example, the determination of the experimental stability ratio for the native
particles tabulated in Table 5—2. At high electrolyte concentrations (>002 M KCl) the d

eviation in the stability ratio is small, and so it is easy to calculate a reliable stability ratio.

Table 5-2: Typical deviation in experimental stability ratio

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

electrolyte Experimental stability ratio
concentration 50—1000 50—2000 , 50—3000 50—3600 actual
0.0005 129.0 175.0 101.0 158.0 889.3
0.001 70.0 601.0 296.0 7912.0 432.7
0.002 57.0 1 13.0 95.0 101.3 323.6
0.0045 194.1 81.1 97.3 82.8 118.9
0.91 24.4 66.9 100.3 82.0 66.8
0.02 121.8 131.8 119.8 142.3 32.0
0.045 12.7 22.4 47.9 61 .2 24.9
0.1 27.0 44.8 5.8
0.2 4.2 3.1 2.9 2.6 2.0
0.45 2.2 2.0 1.7 1.5 1.6
0.75 11.8 1 .5 1 .4 1.4 1 .4
slope 068 -0.61 -0.85 -0.77 -0.93

 

 

However, at smaller electrolyte concentrations, the deviations can become large, depending

on what segment of the data is used to obtain a regression line.

Table 5-2 shows an example of the variation that is possible in the determination of the
experimental stability ratio. This is a slightly exaggerated example because no care was
taken in selecting the “smoo ” regions of the intensity profiles. However, it does illustrate
the point that variation exists depending on what region of the intensity profiles are used For

example, at 0.001M KCl, using the intensity measurements fi'om 50-1000 scans gives a

44

stability ratio of 70;. An increase of two orders of magnitude is observed, however, if the

data from 50-3000 scans is used The slope of the curves is affected too. Depending on the

area of the intensity measurements used, the slope varies from -0.6 to -0.9.

5. 7 COMPARISON OF THEORETICAL AND EXPERIMENTAL STABILITY [arms

The theoretical and experimental stability ratio are best compared by looking at the slope of a

line drawn through the stability ratio in the slow flocculation regime.

To aid in this

comparison, shows the slope of all of the curves as well as a ratio (E/I') between

experimental (E) and theoretical (T) slopes. The higher this ratio, the closer the agreement

between experiment and theory. A value of unity would indicate that the

Table 5-3. Comparison between theoretical and experimental stability ratios

 

 

 

 

 

 

 

 

 

Theoretical CF) Experimental (E) Ratio (1:71)
Stabilized -15.41 -1.77 .115
24 hour heat treated stabilized -12.18 -1 .15 .094
Native -14.09 -0.96 .068
24 hour heat treated native -12.87 -0.549 .043

 

experimental and theoretical curve are in complete agreement Based on this value, the order

of agreement between experiment and theory is as follows: stabilized (Figure 5-9b), 24 hour

stabilized (Figure 5-9d), native (Figure 5-9a) and 24 hour heated stabilized (Figure 5-9b).

 

 

45

 

 

 

 

 

 

 

 

 

 

 

 

 

‘. 1000 3 H\‘ 1 1000.000 1
' 1 \e 1 Z
‘ 2 j ‘. .- .I 1 r
; 2 1 . z ‘\ ‘ ‘
i ‘1 100 .- ' 1 ‘ 1 100.000 3
Q\ i 1 g . “EN"... \ j
1 “' 1 1 = ‘~ \
g . \ o 1 1 1‘ *1“‘~~.ag .
\ 10 ' \ a 10000
. T‘\ 3 \\ \\ a I
i \ \~. .\ El? -1
' \ h g."
I: . 1 \. g 1 T I r r 1 fr 1 1 1 $1‘%
0.01 0.1 0-01 0-1
electrolyte concentration, M ? electrome concentration. M
1 . Emenmenta' nathe 3 Emenmental Stablllmd
__,.____ Theoretical native —— Theoretieal Stabilized

 

-.....J

 

 

 

(a)

 

 

(b)

 

 

 

 

1 -- --'iheor. 24 hr. ht. treated

 

 

 

1

 

 

(C)

, Theor. 24 hour ht. treated stab.
l—

 

 

‘. i 3
t 3 1
. 1 100 3* ix 100 3
f l '1 ' \\ ‘ 3
' l ' 1 “ \\
g \ " 1 a ; \\ . 1
A.\“-\ I ‘ l \\ “
. ‘\\\..‘, 10 1» fifi 10 .1.
I \ \
f \\ ~ ‘_ q‘ 1 ‘ ‘\\A 1
1 T » r I . Hi T r. 11' 1 v r 1 . r r l . r 11. r
0.01 0.1 1 0,01 0,1 1
electrolyte concentration, M electrolyte concentration, M
1 ... Expt. 24 hr. ht. treated "

 

a Expt. 24 hr. ht. treated stab.

 

((1)

Figure 5-9: Stability Ratios of Native and Stabilized Particles.

 

46

The improvement between theory and experiment upon the addition of a surfactant is not
surprising because it has been reported in the literature that such an improvement occurs in
deposition studies (Kayes, Litton and Olsen, 1995). However, the reduction in improvement
was a little startling. The ratio Efl‘ decreased with heat treating from 0.115 to 0.094 for the
stabilized particles and 0.068 to 0.043 for native particles. Many studies exist that point out
the improvement upon heat treating (Rosen and Saville, 1991; Ofoli, 1994; Elimelech and

O’Melia, 1990).

To see why this might be so, a closer look at the data is warranted (Table 5-4). For
comparisons between theory and experiment, it is wise to use the same range of
concentrations. A look at the data, however, shows that data was not collected over the same
concentrations ranges to result in the same comparison I, therefore, had to resort to

comparing over the same range of stability ratio, which might not be as accurate.

Another possibility for the disagreement with other researchers lies in the analysis of the
experimental stability ratio. To aid in a direct comparison between curves, all electrolyte
concentrations were normalized by their individual 000 in Figure 5-10. It is interesting to
note that the theoretical (T) stability ratios are basically the same for all data represented
The only variation comes from the experimental stability ratio. Therefore, any error in
calculating the experimental stability ratio would detract from the improvement between

theory and experiment.

47

Table 5-4: Unstabilized particles

 

48

 

 

 

 

 

 

 

 

 

. . D 0 Native 1

. o 24H Nat

. -.-

Q o . g 100 a Stab. ;
e ’ o . o 24H Stab:

. ”a 8 ————Native-T

. ° D 10 - Stab-T
: ~24l-IN-T ‘
3 8 24Hs—T

1.E-03 1.E-02 1.E-01 1.E+00 1.E+01

Normalized Electrolyte Concentration, c/ccc

 

 

Figure 5-10: Normalized stability ratio

 

 

6. SUMMARY AND CONCLUSIONS

6.1 SUMMARY

This purpose of this work was twofold: to examine the effect of masking surface charge
heterogeneities with a surfactant on the discrepancy between theoretical and experimental
stability ratios and the effect of heat treating the surface of the colloid on that same disparity.
Sulfate stabilized, surfactant free, polystyrene latex particles were used as a model colloid

since they are monodisperse with well characterized surface functional groups.

Heat treatment of the particles has been shown to improve the agreement between
experiment and theory (Rosen and Saville, 1990; Ofoli, 1994). It was achieved in this study
by heating a 0.05 vol% fraction of the particles for 6, 12 and 24 hour at 120°C followed by a
slow cooling. The resulting dispersions were then tested by low angle light scattering and

electrophoresis to get an experimental and theoretical stability ratio.

Surface charge heterogeneities on the colloid have been shown to affect the theoretical
stability ratio (Kayes, 1990). Litton and Olsen (1995) showed that adding a surfactant to a
polystyrene dispersion reduces the discrepancy between theory and experiment. Therefore, a
surfactant was added to some of the dispersions to achieve a final concentration of lxlO’5 M
SDS. Surfactant was added to the heat treated dispersions only after the heat treatment
regime was completed These samples were also tested by light scattering and

electrophoresis to obtain an experimental and theoretical stability ratio.

49

50

The light scattering setup used was that described by Prieve and Young (1991) with the
intensity of the scattered light measured at l.6°. Electrophoresis was completed on a Penn
Kern Model 501 Lazer Zee instrument Zeta potential was obtained from the electrophoretic

mobility using the Smoluchowski equation.

The experimental stability ratio was determined by obtaining a light scattering intensity vs.
time plot from the small angle light scattering apparatus. A regression, line is then drawn
through data to obtain the slope and intercept of the data These values, as well as the
temperature and the initial singlet concentration of the dispersion, are then substituted into
equation 6. The theoretical stability ratio was obtained by using the equation pmposed by
Prieve and Lin (1982) which takes into account variations in size and surface charge. The
Hammaker constant was obtained from Lifshitz theory using no adjustable parameters. The

standard deviation was also incorporated into the calculations using a MathCad program

designed by Ofoli.

The data show that there is a decrease in overall stability upon heat treatment while the
addition of a surfactant serves to increase the stability of a system which is in agreement with
data reported by other researchers. However, this study did not show the increase in
agreement between theory and experiment with heat treatment that tars been reported by so

many other researchers.

51

6.2 CONCLUSIONS

Heat annealment of the particles did not improve the agreement between theory and
experimental values. However, the experimental stability ratios of the stabilized colloidal
dispersion were in better agreement with theory. The decrease in stability ratio seen upon
heat treatment and the increase observed upon stabilization of the dispersions were in line

with those reported in the literature.

It was surprising, however, that heat annealment did not improve the discrepancy between
theory and experiment as has been reported by other researchers. A possible explanation for
this is that the distribution in zeta potentials of the polystyrene particles had a higher standard
deviation than the 15% assumed Ofoli (1994) showed that taking the deviation of zeta
potential into account greatly reduces the slope of the theoretical stability ratio curve.
Therefore, if the standard deviation in zeta potential of the dispersions used was larger than
15%, the slope of the theoretical heat treated native stability ratio curve would be lower than

that reported

In a similar vein, if the effect of stabilizing the particles is to mask the surface charge
heterogeneities, then the deviation for the stabilized particles would be lower than that for the
native particles. This would increase the slope of the theoretical stability ratio curve for the
stabilized particles, thereby decreasing the agreement with experiment. Further studies
should be done using an electrophoresis instrument capable of measuring the deviation in

zeta potential so that a careful analysis of the standard deviation can done.

7. SUGGESTIONS FOR FUTURE RESEARCH

Exploring the effect of stabilization in more detail would be enlightening Ifthe reduction in
discrepancy seen in this study is real, then the possibility exists for even greater reduction at
higher SDS concentrations. However, further studies should deal with polystyrene particles
stabilized with sulfonate groups as they do no hydrolyze on heat treating. SDS is also subject
to hydrolysis, therefore, another surfactant such as Cetyl Triammonium Bromide or sodium

dodecyl sulfonate should be used

52

APPENDICES

 

APPENDIX A - STOCK SOLUTION CALCULATIONS

 

1. APPENDIX A: STOCK SOLUTION CALCULATIONS

Table 8-l: Concentration calculator for stock colloid

 

 

 

 

 

 

 

 

 

 

 

Stock Colloid

MW WWII FlndVblumumlnconcumdadnd nrltoadd

8.8 as 250 0.05 1.42E+00

5.87E+12 particleslmL 250 I 3.33810 1.42900

Table 8-2: Calculator for stock SDS solution
I SDS Solution

| MW units Vina! Volume (mlbncentratlon desire g to add

| 288.38 g/mol 50 2.219E-04 0.0032

 

 

 

 

 

 

Table 8-3: Calculator for high molarity KCl stock solution

 

KC! calculator

Desired Molaity: m
ml. of solution: 100.00

cdculahd g KCL: 22.35

 

 

Actual Mommy: m
mL of solution: 100.00

actual g KCL: 22.3554

 

 

53

 

 

54

Table 8-4: Calculator for low molarig/ KCl stock solution

 

KCl calculator

Desired Molarity: ‘11-:

mL of solution:

calculated 9 KCL:

100.00

1.11825

 

 

Actual Molarity: 0.1507

mL of solution:

actual g KCL:

100.00

1.1234

 

 

Table 8-5: Worksheet used for typical colloidal dispersions

 

The unstabilizedpaflicles were prepared directly from stock colloid

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

wine! cone. units T Final Vol.(ml) conc. desired ml to add

3.35+10 particlesImL 50 2.345+09 3.51

Sample as run

wine! conc. units Final Vol.(ml) conc. desired ml to add

2.3E+09 particles/mL 3 7.8E+08 1 .0

Concentration Calculator for final samples

Electrolyte concentration? 0.1507 M

mLKCl mL water ml particles total volume final cone.
0.002 1.998 1.000 L 3.000 1.005-04
0.010 1.990 1.000 3.000 5005-04
0.020 1.980 1.000 g 3.000 1.005-03
0.050 1.950 1.000 L_ 3.000 2.505-03
0.100 1.900 1.000 g___ 3.000 5.005-03
0.149 1.851 1.000 .N 3_ 3.000 7.505-03
0.199 1.801 1.000 ,__ 3.000 1.005-02
0.249 1.751 1.000 ' 3.000 1.255-02
0.299 1.701 1.000 ; 3.000 1.505-02
”0.348 1.852 1.000 L 3.000 1.755-02
0.995 1.005 1.000 T 3.000 5005-02
2.000 0.000 _ 1.000 ‘ 3.000 1.005-01
0.817 1.183 ___ 1.000 f 3.000 4.50501
1.362 0.638 1.000 3.000 7.505-01

 

 

APPENDIX B

 

Seamed 5:53qu 88 30393 “To Page

     
         
 

   

  

doe. 0:. 9.2.2»
0.20.. 0.00» «13-x .ow

 

.....nan..

.... 1.14.5.1. 1
in... .111... .

 

 

 

 

 

 

 

 

 

 

:—
m. o
o
3. a

_ . E 0:. .o 0:: .2: 2:
.... 8 2.: sec» 2.: 2.:-
. .. .p f . {5.41 ...q...........w..... n............ 3.... 1 J.” .5 1...:...- ........... ... . .....u ‘0‘. ngcmflhmagfl Us:
.3305 um :25; no.» 5 3 «500° .23 . :3» can 050:5“.

S as. ._ ...... e5 as 2.. 2 ._ «5.. .3...
$2.333» e 2 s .3230 .33 0... tea:

3H3? "m X—GZm—m: .m

55

 

APPENDIX C

1. APPENDIX C: RESULTS

Table 10-1: Stability ratio data sheet for native colloidal dispersion

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Unstabilized native
Experimental Theoretical

conc. W regr. W W
0.0005 889.3 1034.1
0.0010 432.7 531.2
0.0020 323.6 272.9
0.0045 118.9 125.2
0.010 66.8 58.1
0.020 32.0 29.9
0.030 2.615+07
0.045 24.9 13.7
0.050 4811.0
0.052 2601.0
0.053 1645.0
0.055 951.7
0.057 547.8
0.058 438.2
0.059 321.1
0.061 191.4
0.063 118.1
0.066 51.0
0.068 32.8
0.070 21.9
0.075 8.2
0.081 3.9
0.085 2.8
0.093 2.0
0.100 5.8 6.4
0.103 1.9
0.189 1.9
0.200 2.0 3.3
0.450 1.6 1.5
0.750 1.4 0.9

 

 

 

 

 

5t”

' fia‘la .'.-

 

57

Table 10-2: Ratio data sheet for stabilized colloidal '

 

'mental
r. W W
1310.9 1230.7
385.6 367.6
60.3 74.4

 

 

58

Table 10-3: Stabili ratio data for 24 hour heat treated native dis rsion

  

   

 

 

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

24 heat treated

Experimental Theoretical
cone. W regr.w W
0.0001 269.0 255.8
0.0005 98.1 105.7
0.0010 71.2 72.2
0.0020 40.8 49.4
0.0040 48.3 33.8
0.0075 28.0 23.9
0.010 15.7 20.4
0.020 13.1 14.0
0.031 39170.0
0.033 8920.0
0.038 1509.0
0.040 881.7
0.042 314.4 5.
0.044 153.1 I
0.048 79.0 I
0.047 46.1 I
0.049 22.9 I
0.050 8.9 8.4
0.051 18.1 I
0.055 7.1
0.057 5.2 1
0.059 4.0
0.062 2.8
0.076 1.9 I
0.098 1.9
0.196 1.9
0.2 1.9 3.9

 

 

 

 

 

 

Table 10-4: Stabili

ratio data for 24 hour heat treated stabilized colloidal dispersion

S9

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

24 hour heat treated
r Experimental ITheoretical

cone. W regr. W W
0.010 94.380 127.823
0.020 41.970 59.456 ]
0.050 68.822 21.815 9238.00 I
0.055 1821.00
0.057 1177.00
0.058 758.58

I 0.080 439.90
0.081 381.01 I
0.085 124.29 I
0.088 83.55
0.071 34.44

I 0.075 17.02
0.080 8.28 J
0.085 4.79 I
0.095 2.49
0.100 13.848 10.054 2.20
0.125 1.89
0.150 8.271 8.425
0.193 1.89
0.200 2.755 4.877
0.250 2.299 3.655
0.259 1.89 I
0.400 2.750 2.175 I

I 1.980 2.057 0.372 |

 

 

 

 

 

Table 10-5: Stability ratio data for 6 and 12 hour heat treated native and stabilized
colloidal dispersions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6 hour heat treated
cone. W regr. W
0.0001 301.704 215.787
0.0005 100.127 88.768
0.0010 48.718 58.805
0.0020 33.669 39.585
0.0040 21.819 26.737
0.0075 14.873 18.732
0.0100 12.300 15.917
0.0200 11.398 10.751
0.0500 9.238 6.400
0.2000 3.340 2.920

(a) 6 hour heat treated native
6 hour heat treated
cone. W regr. W
_ '

0.01 391.856 398.645
0.02 1 10.385 144.093
0.05 33.812 37.534
0.1 23.892 13.587
0.15 17.307 7.481
0.2 3.910 4.904
0.25 1.804 3.534
0.4 1.638 1.773
1.95 1.783 0.173
(c) 6 hour heat treated stabilized

12 hour heat treated
cone. W regr. W
0.0001 313.261 319.933
0.0005 102.398 139.389
0.0010 102.050 97.460
0.0020 72.126 68.144
0.0040 43.707 47.646
0.0075 49.095 34.442
0.0200 29.744 20.758
0.0500 15.460 12.935
0.1005 5.055 9.023
0.2000 1.780 8.324

(b) 12 hour heat treated native
12 hour heat treated
com w ream
0.01 240297 221.428
0.02 69.994 87.589
0.05 26.115 25.690
0.1 11.285 10.180
0.15 8.492 5.905
0.2 3.314 4.018
0.25 2.792 2.981
0.4 1.463 1.589
1.95 1.765 0.191

 

 

 

 

(d) 12 hour heat treated stabilized

 

 

COl’lC.

0.02979
0. 05769
0.09303

CODC.

0.03072
0.06234
0.09768

61

Table 10-6: Zeta Potentials

COHC.

0.06327
0.09582
0.18696

COl’lC.

0.06141
0.09489
0.19254

 

 

 

BIBLIOGRAPHY

 

 

BIBLIOGRAPHY

Chapman, D.L. Phil Mag, 25, 475 (1913)

Chow, RS. and K. Takamura, Effects of surface roughness (hairiness) on latex
particles on their electrokinetic potentials. J. Colloid and Interface Sci, 125: 226
(1988)

Derjaguin, B. V. and Landau, L. Theory of the stability of strongly charged lyophobic
sols and of the adhesion of strongly charged particles in solutions of electrolytes.
Acta Physioclrim URSS. 14: 633 (1941).

Einarson, MB. and Berg, J .C. Electrostatic stabilization of colloidal latex dispersions. J.
Colloid Interface Sci. 155: 165 (1993).

Elimelech, M. and C. O’Melia. Colloids Surfaces, 44: 165 (1990).
Gouy, G. J. Phys, 9: 457 (1910)
Gregory, J. and AJ. Wishart, Colloids Surf, 1: 313 (1980).

Hidalgo—Alvarez, R., J.A. Molean, R]. de las Nieves and B.H. Bijsterbosch, J. Adv.
Colloid Interface Sci., 149: 23 (1992).

Hidalgo-Alvarez, R., Martin, A., Fernandez, A, Bastos, D., Martinez, F. and de las
Nieves, F .J ., Electrokinetic properties, colloidal stability and aggregation kinetics
of polymer colloids. Advances in Colloid and Interface Science, 67: l (1996)

Kerker, M, The scattering of light and other Electromagnetic radiation. (Academic Press,
New York, 1969), chap 8, p 414.

62

63

Kihara, H., N. Ryde, and E. Matijevic, Kinetics of heterocoagulation Part 2-The effect of
discreteness of surface charge. J. Chem. Soc. Faraday Trans. 88: 2379 (1992).

Kihara, Hiroshi and Matijevic, Egon. Kinetics of Heterocoagulation 3. Analysis of
effects causing the discrepancy between theory and experiment. Langmuir, 8:
2855(1992)

Laidler, K]. and Meiser, 1.11. Physical Chemistry. The Benjamin/Cummings Publishing
Company, Menlo Park (1982)

Lips, A. and Willis B. Low angle light scattering technique for the study of coagulation.
J. Chem. Soc. Faraday Trans. 69: 1226 (1973).

Litton, GM. and Olson, TM. The influence of Particle Size on Colloid Deposition
Kinetics with Sodium Dodecyl Sulfate. Submission to Colloids and surfaces
(1995).

Myers, DR and Saville, D.A. Dielectric spectroscopy of colloidal suspensions II.
Comparisons between experiment and theory. J. Colloid Inted'ace Sci. 131: 461
(1989).

O’Brien, R.W. and White, L.R. Electrophoretic mobility of a spherical colloid particle.
Chem. Soc. Faraday Trans. 11 72: 1607 (1978).

Ofoli, KY. and Prieve, D.C. Kinetics of latex flocculation: effect of annealing and a
distribution in surface potential. J. Colloid Surface Science, submitted.

Ofoli, Robert Y. Initial Rate Of Flocculation Of Polystyrene Latexes: Effects Of Surface
Annealing And A Distribution In Surface Potential. Ph.D. Dissertation, Carnegie
Mellon University, Pittsburgh, PA (1994).

Oster, G. and DP. Riley, Scattering From Isotropic Colloidal And Macromolecular
Systems. Acta Cryst, 1952, 5,1.

Ottewill, RH. and Shaw, J.N. Stability of monodisperse polystyrene latex dispersions of
various sizes. Disc. Faraday Soc. 42: 154 ( 1966).

 

[P (1:1
1

64

Overbeek, J. Theodoor G, Strong And Weak Points In The Interpretation Of Colloid
Stability. Advances in Colloid and lnterfacial Science. 16: 17 (1982).

Pailthorpe, BA. and Russel, W.B. The retarded van der Waal interaction between
spheres. Colloid Interface Sci. 89: 563 (1982).

Prieve, DC. and Lin, M.M.J. The Effect Of A Distribution In Surface Properties On
Colloid Stability. J. Colloid Interface . 86: 17(1982).

Reerink, H. and Overbeek, J.Th.G. The rate of coagulation as a measure of the stability
of silver iodide sols. Disc. Faraday Soc. 18: 74 (1954).

Rosen, LA. and Saville, D.A. Dielectric spectroscopy of colloidal dispersions:
comparison between experiment and theory. Langmuir 7: 36 (1991)

Rosen, LA. and Saville, D.A The dielectric response of polystyrene latexes: effects of
alterations in the structure of the particle surface. J. Colloid Interface Sci. 140:
82 (1990).

Rosen, LA and Saville, DA The electrokinetic response of surface modified polymer
latexes: effect of grafted water soluble polymer and heat treatment 149N2: 542
(1992)

Russel, W.B., Saville, DA. and Schowalter, W.R. Colloidal Dispersions. Cambridge
University Press, Cambridge, UK (1989).

Shaw, D.J. Introduction to Colloid and Surface Chemisgy, Butterworths (1966).
Smoluchowski, M. Phys.Z l7:557,585 (1916).

Spielman, L.A. Viscous interactions in Brownian coagulation. J. Colloid Interface Sci.
33: 562 (1970).

Stem, 0., Elektrochem. 30: 508 (1924)

Verway, E. J. and Overbeek, J. T. G. Theory of the stability of lyophobic colloids.
Elsevier, Amsterdam (1948).

Von Helmholtz, H.L.F. W iss Abhandl. Physik-tech Reichsanstalt, l: 925 (1879).

 

65

Young, W. D. and Prieve D. C. An Improved light scattering technique for measuring
the flocculation rate of colloids. Langmuir 722887 (1991).

Kayes, 1.8. The effect of surface active agents on the microelectrophoretic properties of a
polystyrene latex dispersion. Journal of Colloid and Interface Science 56: 426
(1976)

Kandori, K., Ishiguro, H., Kon-no, K., and Kitahara, A., Effect of pH and Ionic
surfactants on ionic latices with high surface charge density. Langmuir 5: 1258
(1989)